Our data suggest that the principal clinical and biochemical manifestations of melancholic depression reflect concomitant activation of the two major stress-responsive neurotransmitter systems in brain that had escaped their usual glucocorticoid-mediated counter-regulation. Specifically, we have shown that the organized state of anxiety that is the cardinal clinical manifestation of melancholia occurs in the context of the activation of the corticotropin releasing hormone (CRH) and locus-ceruleus norepinephrine (LC-NE) systems, and that indices of their activation correlate positively with one another. In this regard, our basic studies indicate that the CRH and LC-NE systems participate in a mutual reverberatory positive feedback loop, in which each reinforces the other's functional activity. As an adjunct, we conducted a series of basic studies implicating an important role for CRH in sensitizing key substrates in brain, and hence, in influencing the natural history of recurrent affective illness. Moreover, we showed that the molecular effects of chronic but not acute antidepressant medication involve a consistent decrease in the expression of the CRH gene in the hypothalamus and of the TH gene in the locus ceruleus, in association with an increase in the expression of type 1 glucocorticoid receptor in the hippocampus, thought to be a principal restraining element of the CRH system. Hence, these data are compatible with our idea that melancholia represents an activation of the generalized stress response that has escaped its counter-regulatory influences. We also showed that while the hypercortisolism of major depression represented a defect at or above the hypothalamus resulting in the hypersecretion of CRH, the hypercortisolism of Cushing's disease represented inadequate glucocortoid restraint upon the pituitary. Our data showing suppression of the hypothalamic CRH neuron in patients with Cushing's disease clarified the mechanism of the post-operative adrenal insufficiency in this disorder, provided the rationale for therapeutic interventions, and established the hypothesis that the atypical depression seen in Cushing's disease could reflect a pathological inactivation of the CRH neuron. Our basic studies elucidating the neurotransmitter, neuropeptide, cytokine, and feedback modulation of the central component of the hypothalamic-pituitary-adrenal axis will hopefully contribute to the development of new antidepressant agents, the treatment of steroid-induced adrenal insufficiency, and the treatment of stress-induced hypothalamic amenorrhea and infertility occurring as a consequence of a shortened luteal phase.